# Vals Index ![Category](https://img.shields.io/badge/Category-Composite%20Agentic-1f6feb) ![Metric](https://img.shields.io/badge/Metric-Weighted%20Index%20(%25)-2da44e) ![Saturation](https://img.shields.io/badge/Saturation-Low-d29922) ![Created](https://img.shields.io/badge/Created-2026-555) [![Leaderboard](https://img.shields.io/badge/Leaderboard-vals.ai%2Fbenchmarks%2Fvals__index-0969da)](https://www.vals.ai/benchmarks/vals_index) > **Quick answer:** The Vals Index is Vals AI's composite benchmark measuring AI's potential economic impact: agentic model performance across finance, coding, legal and tax tasks, weighted by each sector's share of U.S. GDP. As of October 26, 2026, [Claude Sonnet 5.5](/models/anthropic/claude-sonnet-5-5) ranks first at 70.85%, followed by [Claude Opus 5.5](/models/anthropic/claude-opus-5-5) (70.18%) and [Gemini 4 Argon](/models/google/gemini-4-argon) (69.24%). ## What is the Vals Index? The Vals Index is a proprietary composite index from Vals AI that rolls up agentic model performance on finance, coding, legal and tax work into a single number, weighting each underlying task category by that sector's share of U.S. GDP. It is positioned by Vals AI as "a single measure of AI's potential economic impact." As of the October 26, 2026 update, 28 models have been tested. ## Benchmark Details | Property | Value | |---|---| | Composition | Finance, coding, legal, tax (agentic tasks) | | Weighting | By each sector's share of U.S. GDP | | Models tested (as of Oct 2026) | 28 | | Test set | Private (Vals AI proprietary) | | Created by | Vals AI | | Leaderboard | [vals.ai/benchmarks/vals_index](https://www.vals.ai/benchmarks/vals_index) | ## State-of-the-Art Results Vals AI leaderboard, updated 2026-10-26: | Rank | Model | Score | Source | Date | |---|---|---|---|---| | 1 | [Claude Sonnet 5.5](/models/anthropic/claude-sonnet-5-5) | 70.85% | [Vals AI](https://www.vals.ai/benchmarks/vals_index) | 2026-10 | | 2 | [Claude Opus 5.5](/models/anthropic/claude-opus-5-5) | 70.18% | [Vals AI](https://www.vals.ai/benchmarks/vals_index) | 2026-10 | | 3 | [Gemini 4 Argon](/models/google/gemini-4-argon) | 69.24% | [Vals AI](https://www.vals.ai/benchmarks/vals_index) | 2026-10 | ## Vals Index on Benchgen > No Benchgen runs. The test set is private and run exclusively by Vals AI, so the scores above are mirrored from Vals AI and cannot be re-run on Benchgen. ## Frequently Asked Questions

What is the Vals Index?

The Vals Index is Vals AI's composite benchmark combining agentic model performance across finance, coding, legal and tax tasks into one score, weighted by each sector's share of U.S. GDP, framed as a single measure of AI's potential economic impact.

Which model leads the Vals Index?

As of October 26, 2026, Claude Sonnet 5.5 leads at 70.85%, ahead of Claude Opus 5.5 (70.18%) and Gemini 4 Argon (69.24%).

--- *Source: [Vals AI benchmarks](https://www.vals.ai/benchmarks). Last updated 2026-10-27.*

Rank	Model	Score
1	claude-sonnet-5-5	70.85
2	claude-opus-5-5	70.18
3	gemini-4-argon	69.24

Vals Index — Results

Vals Index — Evaluations